The full upstream README, mirrored here for reference. Install config, tool schemas, adoption signals, and an original overview live on the Github listing page.
An MCP (Model Context Protocol) server that provides GitHub analytics and management tools for AI agents. Analyze repositories, summarize pull requests, triage issues, generate release notes, and track contributor activity — all through a standardized tool interface.
Analyze a GitHub repository and get a comprehensive overview including stars, forks, open issues, language breakdown, top contributors, and weekly commit activity trends.
Summarize a pull request with diff stats, categorized file changes, review status (approvals, change requests, pending reviewers), and recent comment excerpts.
Analyze open issues in a repository. Automatically categorizes issues by labels and keywords (bug, feature, docs, security, etc.), suggests priorities based on engagement and age, identifies potential duplicates using text similarity, and flags stale issues.
Generate release notes from commits between two references (tags, branches, or SHAs). Categorizes commits using conventional commit patterns and produces both Markdown and structured JSON output.
Get contributor statistics within a configurable time window: commits, PRs opened/merged, code reviews, and identification of first-time contributors.
Add to your Claude Desktop configuration file:
Windows: %APPDATA%\Claude\claude_desktop_config.json
macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Add the server using the CLI:
Or add it to your .claude/settings.json:
Once connected, the tools are available to the AI agent automatically. Here are example prompts:
All tools accept an optional token parameter for authenticated requests. Without a token, the GitHub API allows 60 requests per hour. With a token, the limit increases to 5,000 requests per hour.
You can create a personal access token at github.com/settings/tokens. No special scopes are required for public repository access.
MIT